Online Survey & Research Panelist — Various Platforms (Prolific, Remotasks)
Performed text labeling and annotation tasks for AI research projects using sentiment classification, text ranking, and instruction evaluation. Applied structured guideline-based judgment to assess outputs and support downstream model training. Provided consistent ratings reflecting quality dimensions required by the task instructions. • Sentiment classification on provided text inputs • Text ranking comparisons to establish ordering/relevance • Instruction-following evaluation for response quality • Quality assurance to maintain high acceptance rates